A joint frequency-based plant pollutant distribution hotspot prediction method

Abstract

The present application relates to the technical field of fluid-structure coupling simulation calculation, and particularly relates to a plant pollutant distribution hotspot prediction method based on joint frequency. The core innovation of the method is as follows: 1) a simulation model establishment and calculation parameter setting whole process method suitable for different sites is constructed, so as to ensure the calculation stability and the applicability of different sites; 2) a parameterized scanning setting method based on joint frequency and a parameterized scanning program independently developed based on python are proposed, so as to solve the problems of too many calculation groups in parameterized scanning and difficult manual setting, and realize automatic analysis of joint frequency, parameter setting, calling operation and storage; 3) a multi-natural-condition data matrix visualization program is independently developed based on python, so as to break through the bottleneck of long-term pollutant deposition prediction and analysis at the level of month, year and multiple years.

Description

Technical Field

[0001] This invention relates to the field of fluid-structure interaction simulation technology, and in particular to a method for predicting hotspots of pollutant distribution in a plant area based on joint frequency. Background Technology

[0002] The distribution and deposition analysis of airborne effluents emitted from multiple exhaust stacks within the plant area is a crucial issue concerning the plant's environmental safety and the health of its workers. However, variable weather conditions, with continuous changes in wind direction, wind speed, and atmospheric stability throughout the year, pose challenges to the analysis of long-term and annual pollutant deposition.

[0003] Currently, the main methods used for predicting pollutant distribution include traditional methods and computational fluid dynamics (CFD) methods. Traditional methods include empirical formula methods and Gaussian model calculation methods, which are relatively fast, but their accuracy depends heavily on the selection and adaptation of empirical parameters. Generally, only the wind direction and speed with the highest frequency throughout the year are used for conservative analysis. Computational fluid dynamics methods are another mainstream research approach. They mainly use the finite element method to transform the entire fluid region into a Lagrange grid and explicitly solve the Navier-Stokes equations at each node of the grid. Computational fluid dynamics methods are more accurate than traditional methods, but they require a large amount of computational resources and time, and their computational stability is difficult to guarantee, making them unsuitable for the long-term cumulative distribution of pollutants under variable weather conditions.

[0004] Therefore, the current main fluid dynamics simulation methods have difficulty controlling long-term computational stability, require large computational resources, and take a long time to compute, making them unsuitable for direct use in monthly, annual, and multi-year long-term pollutant deposition prediction and analysis.

[0005] This invention proposes a method for predicting pollutant distribution hotspots in industrial plants based on joint frequencies. This method utilizes joint frequencies to globally statistically analyze multiple CFD simulation results, comprehensively covering annual meteorological variations and achieving reliable long-term deposition prediction. Specific steps include model establishment, grid generation and parameter control, calculation parameter adjustment, joint frequency parameterized scanning analysis, weighted superposition calculation, and result visualization analysis. By superimposing pollutant concentration distribution results under multiple grouped natural meteorological conditions using joint frequencies, an annual cumulative hotspot distribution map of the industrial plant is established. This method significantly reduces computational complexity while ensuring comprehensive coverage of variable meteorological conditions throughout the year, achieving efficient and reliable long-term pollutant deposition prediction. This invention effectively solves the technical difficulties in long-term cumulative environmental risk analysis of industrial plants and is of great significance for environmental protection and human health risk analysis. Summary of the Invention

[0006] The purpose of this invention is to propose a method for predicting the distribution hotspots of pollutants in industrial plants based on joint frequency. The method includes multiple steps such as model building, grid division and parameter control, calculation parameter adjustment, joint frequency parameterized scanning analysis and result visualization analysis. Parallel strategies are set to improve computational efficiency, and it can realize the analysis of monthly, annual and multi-year pollution cumulative effects. It has important practical significance and application value in industrial plant environmental protection and personnel health risk analysis.

[0007] The technical solution adopted in this invention is as follows: The first aspect of this invention provides a method for predicting hotspots of pollutant distribution in a plant area based on joint frequency, comprising the following steps: (1) Setting of fluid dynamics calculation parameters: Based on the geometric model of the plant area, mesh generation, numerical model construction and boundary condition setting are carried out to establish a numerical simulation model; (2) Joint frequency parameterized scanning: Using a parameterized scanning program, automated fluid dynamics simulation calculations are performed on various combinations of wind direction, wind speed and atmospheric stability. The calculations are performed in parallel. (3) Results visualization processing: The pollutant concentration distribution data obtained under different natural conditions are superimposed with the joint frequency as the weighting coefficient to generate monthly, annual or multi-year cumulative pollution hotspot area distribution maps; For a calculation point in the region, the joint frequency weighted superposition formula is as follows: Among them, C total The final concentration after weighted summation at a certain calculation point; c i Let ν be the concentration at a certain calculation point under a certain joint frequency meteorological condition. i The corresponding frequency value under the joint frequency meteorological conditions. The joint frequency mentioned in this invention is based on meteorological statistics for a specific factory area, representing the probability of simultaneous occurrence of three meteorological conditions: wind direction, wind speed, and atmospheric stability. In some preferred embodiments, the numerical model construction in the fluid dynamics calculation parameter settings includes a turbulence model, which adopts the k-ε model. Its turbulent kinetic energy k-transport equation and turbulent dissipation rate ε equation are as follows: Turbulent kinetic energy k-transport equation: ; Equation for turbulent dissipation rate ε: .

[0008] In some preferred embodiments, the fluid dynamics calculation parameter settings include a height-dependent fluid inlet velocity, an inlet condition setting where the wind direction is perpendicular to the inlet surface, and a fluid velocity set as a height-dependent function. ; Among them, U 10 The annual average wind speed at a height of 10 m; p This is the wind profile index.

[0009] In some preferred embodiments, during the joint frequency parameterization scanning step, the atmospheric stability is classified into six categories: A, B, C, D, E, and F, corresponding to different wind profile indices and vertical temperature gradient values. The temperature gradient formula is as follows: ; Where T is the temperature at any height, T0 is the temperature at the ground, and z is the vertical height from the ground. This represents the vertical temperature gradient.

[0010] In some preferred embodiments, during the fluid dynamics calculation parameter setting step, a rare substance outlet boundary condition is set for the exhaust surface of the plant exhaust stack, and the substance diffusion follows the convection-diffusion equation: , ; Where J is the diffusion flux, kg / (m²) 2 ·s), where u is velocity, m / s, Here, C is the Hamiltonian operator, and C is the concentration in mol / m³. 3 R is the source term, mol / (m 3 ·s), where D is the diffusion coefficient, m 2 / s.

[0011] In some preferred embodiments, in the joint frequency parameterization scanning step, the wind direction is set to 16 directions, and the atmospheric stability is set to six categories: A, B, C, D, E, and F. The program automatically skips meteorological condition combinations with a joint frequency of zero to reduce the total number of calculation groups.

[0012] In some preferred embodiments, during the hydrodynamic computational parameterization scan step, the SIMPLE solver is used for steady-state solution and parallel computation is enabled.

[0013] A second aspect of the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, implements the method for predicting hotspots of pollutant distribution in a plant area based on joint frequency as described in any one of claims 1 to 7.

[0014] A third aspect of the present invention provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method for predicting hotspots of pollutant distribution in a plant area based on joint frequency.

[0015] A fourth aspect of the present invention provides a hotspot prediction system for pollutant distribution in a factory area, comprising: Data acquisition and modeling module: used to acquire geographical information of the factory area, building data and exhaust stack parameters, and to build a three-dimensional geometric model; Parametric simulation calculation module: configured to perform the joint frequency parameterization scan step of the above method; Visualization and Analysis Module: Used to generate and display maps of pollution hotspot areas and perform risk analysis.

[0016] The technical solution adopted in this invention can achieve the following beneficial effects: This invention proposes a method for predicting the distribution of pollutants in a factory area based on joint frequency. The core innovation of this method lies in: 1) A complete process method for establishing simulation models and setting calculation parameters applicable to different sites was constructed to ensure calculation stability and applicability to different sites; 2) A parameterized scanning setting method based on joint frequencies and a parameterized scanning program developed independently in Python were proposed, which solved the problems of a large number of calculation groups and difficulty in manual setting of parameterized scanning, and realized automatic analysis, parameter setting, calling calculation and storage of joint frequencies; 3) Based on Python, we independently developed a visualization program for multi-natural condition data matrices, breaking through the bottleneck of monthly, annual, and multi-year long-term pollutant deposition prediction and analysis.

[0017] This invention effectively solves the technical difficulties in analyzing long-term cumulative environmental risks in factory areas, and is of great significance for environmental protection and personnel health risk analysis. The difference between this invention and fluid mechanics methods lies in its use of joint frequency statistics and parametric scanning. It predicts the annual cumulative distribution by weighting and superimposing results under multiple meteorological conditions, emphasizing efficient statistical integration and computation, and focusing more on global statistical optimization. This invention effectively solves the technical difficulties in analyzing long-term cumulative environmental risks in factory areas, and is of great significance for environmental protection and personnel health risk analysis. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below, forming part of the present invention. The illustrative embodiments of the present invention and their descriptions explain the present invention and do not constitute an improper limitation of the present invention. In the accompanying drawings: Figure 1 This is a flowchart of the combined frequency pollution prediction and analysis method of the present invention.

[0019] Figure 2 This is a flowchart of the algorithm for automated scanning simulation in this invention.

[0020] Figure 3 This is a flowchart of the algorithm for the visualization program of this invention. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. In the description of this invention, it should be noted that the term "or" is generally used to include the meaning of "and / or," unless otherwise expressly indicated.

[0022] Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0023] Example 1 (1) Setting of fluid dynamics calculation parameters Following the methodology of "geometric modeling - mesh generation - numerical model construction - boundary conditions and solution settings," a numerical simulation model suitable for different plant areas is established. See details... Figure 1 The main steps include: ① Using the plant's CAD planning map and elevation data, the plant environment and facilities were geometrically modeled using the 3D modeling tool Rhino. The modeling area was a cuboid with a range of 2.5 km and a height of 1 km, centered on the geometric center of the plant site. Topographic data of the plant area and surrounding areas was downloaded using ArcGIS software to generate elevation files, which were then imported into the model as parametric surfaces to ensure the plant building model matched the terrain. After the plant modeling was completed, the plant model and the edges of the rectangular area were rotated to ensure the plant model always maintained the same orientation, with the edges of the rectangular area facing the incoming flow direction, facilitating subsequent visualization of boundary conditions and concentration accumulation.

[0024] ② Use ICEM or Fluent Mesh software to mesh the geometric model. The mesh should primarily consist of tetrahedral meshes, with a minimum mesh size of 8 m. Apply a mesh refinement boundary layer to the buildings and ground surfaces, setting the number of boundary layers to 5 and the boundary layer growth factor to 1.2. Control the overall uniformity of the volumetric mesh using surface meshing, ultimately ensuring that the overall average mesh quality is above 0.6.

[0025] ③ The numerical model construction mainly includes four parts: fluid control equations, turbulence model, wall function, and mass diffusion model. Fluid control equations: Fluid flow follows three fundamental conservation laws: the law of conservation of mass, the law of conservation of momentum, and the law of conservation of energy. The governing equations are the mathematical descriptions of these three laws, as follows: Mass conservation equation (continuity equation): In the formula: ρ and U are the fluid density and velocity vectors, respectively.

[0026] Momentum conservation equation: In the formula: u, υ, w, μ, and p are the velocity components, dynamic viscosity, and pressure of the fluid in the three directions, respectively.

[0027] Energy conservation equation: In the formula: T, λ, Φ, and Sh are the fluid temperature, thermal conductivity, dissipation function, and internal heat source of the fluid, respectively.

[0028] Turbulence model: The k-ε model was used for turbulence simulation. The k-ε model demonstrates significant advantages in engineering flow simulation, requiring less mesh than Large Eddy Simulation (LES), and exhibiting good prediction accuracy for turbulent structures such as attached flows, jets, and pipe flows. It is particularly suitable for wall-dominated boundary layer flows and channel flows. Processing plant facilities are diverse, with emission sources including elevated and low-level sources; numerous buildings and structures contribute to the complex and variable near-field airflow influenced by buildings and underlying surfaces, resulting in complex migration and diffusion processes of radioactive effluents. The fluid flows in this process are primarily pipe and channel flows, making the k-ε model suitable for simulation.

[0029] Turbulent kinetic energy k-transport equation: Equation for turbulent dissipation rate ε: Turbulent kinetic energy generation term: In the above formula: ρ—fluid density, kg / m³ 3 ; u—velocity, m / s; —Hamiltonian operator: ; p—pressure, Pa; l—turbulent length, m; F — Volume force, N / m 3 ; μ—dynamic viscosity, Pa·s; k—turbulent kinetic energy, m 2 / s 2 ; These are the model coefficients, with typical empirical values ​​of 1.44, 1.92, 0.09, and 1, respectively.

[0030] The reference pressure was chosen as atmospheric pressure (1 atm), and the reference temperature was 293.15 K.

[0031] Wall function: The wall function serves as a bridge connecting the viscous-dominated region near the wall with the turbulent core region. Its core idea is to reduce convergence difficulty by replacing direct solutions for wall flow with semi-empirical formulas. The wall function must satisfy a logarithmic velocity distribution. Where k≈0.41 is the von Kármán constant; C is the wall roughness parameter.

[0032] Relationship between turbulent kinetic energy and dissipation rate: at the wall, k≈0, dissipation rate ,in Let y be the friction velocity and y be the normal distance to the wall.

[0033] Mass diffusion model: Since the compliant emission concentration from factory exhaust stacks is generally low, this invention uses the dilute substance diffusion assumption for calculation. This diffusion model, based on Fick's Law, can accurately describe the spatiotemporal diffusion of pollutants under the influence of complex terrain and building conditions within the factory area. The spatial distribution of pollutants becomes more uniform over time, and the corresponding diffusion equation is the Fick's Law diffusion equation, as follows.

[0034] Turbulence increases the contact area between different regions of a fluid, allowing for more efficient mass transfer. By adding the convection transfer mechanism to the Fick diffusion equation, we obtain the convection-diffusion equation for matter: In the formula, J is the diffusion flux, kg / (m³). 2 ·s), where u is velocity, m / s, Here, C is the Hamiltonian operator, and C is the concentration in mol / m³. 3 R is the source term, mol / (m 3 ·s), where D is the diffusion coefficient, m 2 / s.

[0035] ④ The calculation boundary conditions in this invention are set as follows: Boundary condition settings are mainly divided into two categories: The first category is the overall boundary control of the computational domain. Fluid inlet and outlet conditions are set for two surfaces of the computational domain parallel to the wind direction (upwind inlet surface and downwind outlet surface). The inlet condition is set so that the wind direction is perpendicular to the inlet surface, and the fluid velocity is set as a function related to height. In the formula, U 10 The annual average wind speed at a height of 10 m; p This is the wind profile index.

[0036] The outlet condition on the downwind exit surface is set to a static pressure boundary, with the pressure being the same as atmospheric pressure. A wall boundary condition is applied to the bottom surface (ground) of the computational domain, with the slip condition set to no slip and the wall roughness setting applied. Open boundaries are set for the three surfaces of the computational domain perpendicular to the wind direction (sky, two non-windward sides), allowing airflow to freely enter and exit these surfaces.

[0037] Considering the vertical variation of temperature under different stability conditions, the temperature gradient formula is: In the formula, T is the temperature at any height, T0 is the temperature at the ground, and z is the vertical height from the ground. This represents the vertical temperature gradient.

[0038] The second category involves setting boundary conditions for pollutant emissions from exhaust stacks. Two settings are applied to the exhaust surfaces of the plant's exhaust stacks: fluid velocity setting and rare substance outlet setting. All exhaust surfaces are designated as fluid inlets, with the inlet velocity direction perpendicular to the surface and the inflow velocity uniform across the entire inlet surface. Simultaneously, all exhaust surfaces are designated as rare substance inlets, with the inlet velocity direction perpendicular to the surface and the inflow velocity uniform across the entire inlet surface.

[0039] (2) Joint frequency parameterization scan We used a self-developed Python parametric scanning program to perform automated scanning simulations of multiple parameters (see algorithm flowchart). Figure 2Multi-parameter automated scanning simulation means automating the traversal and simulation of pollutant distribution scenarios with multiple parameter combinations (such as wind direction, wind speed, and atmospheric stability) through programming, avoiding manual calculations. Its purpose is to efficiently generate simulation results covering variable meteorological conditions throughout the year, for subsequent joint frequency weighting analysis to achieve long-term cumulative hotspot prediction. This scan is based on existing joint frequency data, guiding the scanning process (e.g., skipping combinations with a frequency of 0), thereby optimizing computational resources and time. Based on the joint frequency, the scanning parameters control include model rotation angle (wind direction), wind speed, and atmospheric stability. Sixteen main wind directions are set; atmospheric stability is set into six categories: A, B, C, D, E, and F. Each category of atmospheric stability corresponds to its own wind profile index and temperature vertical gradient constant, referencing the urban parameters in GB / T3840-91, as shown in Table 1. Conservatively, a total of 16 wind direction groups, 6 wind speed groups, and 6 stability categories need to be set, i.e., a total of 576 groups. In practice, since some meteorological conditions do not occur in the target plant area (i.e., the joint meteorological frequency value is 0), the simulation group with a joint frequency value of 0 can be automatically skipped in the automated scanning simulation program, reducing the overall calculation time. The total number of calculation groups is generally controlled at around 200-300 groups.

[0040] Table 1. Values ​​of wind profile index and vertical temperature gradient under different stability conditions.

[0041] Subsequently, multiple sets of exhaust stacks and their emission volumes need to be set up according to the actual emission situation of the plant area. Based on the actual emission situation and actual analysis needs, a coupling relationship between emission volume and natural meteorological conditions can be set to improve the accuracy of concentration calculation, but this will also increase the number of calculation sets and the calculation time accordingly.

[0042] In the Fluent fluid dynamics calculation software, select the SIMPLE solver, set the calculation solution type to steady-state solution, and enable the parallel computing strategy to ensure overall operational stability and efficiency.

[0043] (3) Results visualization processing Weighting calculation and result visualization Based on obtaining the pollutant concentration distribution in the factory area under multiple sets of natural conditions, a data point is output every 50 meters, forming a multi-natural-condition data matrix. The specific process of weighted superposition is as follows: for each grid point (data point), the pollutant concentration value under each set of meteorological conditions is multiplied by its corresponding joint frequency weight, and then all sets are summed to calculate the cumulative concentration value of that point, thereby integrating the pollutant distribution throughout the year or multiple time periods. A self-developed Python visualization program is used (see algorithm flowchart). Figure 3The pollution data is superimposed on the base map of the factory area's topography and building distribution, and the weighted superposition of pollutant distribution results is plotted using joint frequency as a weighting coefficient, thereby forming monthly, annual, and multi-year cumulative hotspot area distribution maps.

[0044] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of the present invention.

Claims

1. A method for predicting hotspots of pollutant distribution in a factory area based on joint frequency, characterized in that, Includes the following steps: (1) Setting of fluid dynamics calculation parameters: Based on the geometric model of the plant area, mesh generation, numerical model construction and boundary condition setting are carried out to establish a numerical simulation model; (2) Joint frequency parameterized scanning: Using a parameterized scanning program, automated fluid dynamics simulation calculations are performed on various combinations of wind direction, wind speed and atmospheric stability. The calculations are performed in parallel. (3) Results visualization: Pollutant concentration distribution data obtained under different natural conditions are superimposed using joint frequency as a weighting coefficient to generate monthly, annual, or multi-year cumulative pollution hotspot distribution maps. The joint frequency weighting superposition formula for a certain calculation point in the region is as follows: Among them, C total The final concentration after weighted summation at a certain calculation point; c i Let ν be the concentration at a certain calculation point under a certain joint frequency meteorological condition. i This represents the corresponding frequency value under the combined frequency meteorological conditions.

2. The method according to claim 1, characterized in that, In the fluid dynamics calculation parameter settings, the numerical model construction includes a turbulence model, which adopts the k-ε model. Its turbulent kinetic energy k transport equation and turbulent dissipation rate ε equation are as follows: Turbulent kinetic energy k-transport equation: ； Equation for turbulent dissipation rate ε: 。 3. The method according to claim 1, characterized in that, In the fluid dynamics calculation parameter settings, a fluid inlet velocity related to height is set, the inlet condition is set to a wind direction perpendicular to the inlet surface, and the fluid velocity is set as a function related to height. ； Among them, U 10 The annual average wind speed at a height of 10 m; p This is the wind profile index.

4. The method according to claim 3, characterized in that, In the joint frequency parameterization scanning step, the atmospheric stability is divided into six categories: A, B, C, D, E, and F, corresponding to different wind profile indices and vertical temperature gradient values. The temperature gradient formula is as follows: ； Where T is the temperature at any height, T0 is the temperature at the ground, and z is the vertical height from the ground. This represents the vertical temperature gradient.

5. The method according to claim 1, characterized in that, In the fluid dynamics calculation parameter setting step, a rare substance outlet boundary condition is set for the exhaust surface of the plant's exhaust stack, and the substance diffusion follows the convection-diffusion equation: ， ； Where J is the diffusion flux, kg / (m²) 2 ·s), where u is velocity, m / s, Here, C is the Hamiltonian operator, and C is the concentration in mol / m³. 3 R is the source term, mol / (m 3 ·s), where D is the diffusion coefficient, m 2 / s.

6. The method according to claim 1, characterized in that, In the joint frequency parameterization scanning step, the wind direction is set to 16 directions, and the atmospheric stability is set to six categories: A, B, C, D, E, and F. The program automatically skips meteorological condition combinations with a joint frequency of zero to reduce the total number of calculation groups.

7. The method according to claim 1, characterized in that, In the parameterized scan step of the fluid dynamics calculation, the SIMPLE solver is used to perform steady-state solutions and parallel computation is enabled.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method for predicting hotspots of pollutant distribution in a plant area based on joint frequency as described in any one of claims 1 to 7.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the method for predicting hotspots of pollutant distribution in a plant area based on joint frequency as described in any one of claims 1 to 7.

10. A hotspot prediction system for pollutant distribution in a factory area, characterized in that, include: Data acquisition and modeling module: used to acquire geographical information of the factory area, building data and exhaust stack parameters, and to build a three-dimensional geometric model; Parametric simulation calculation module: configured to perform the joint frequency parameterization scan step of the method of any one of claims 1 to 7; Visualization and Analysis Module: Used to generate and display maps of pollution hotspot areas and perform risk analysis.

Images